“‘Cloze Procedure’: A New Tool for Measuring Readability”, 1953-09-01 (; similar):
Here is the first comprehensive statement of a research method and its theory which were introduced briefly during a workshop at the 1953 AEJ convention. Included are findings from three pilot studies and two experiments in which “cloze procedure” results are compared with those of two readability formulas.
“Cloze Procedure” involves no formula or “element counting”, but consists of sampling all potential readability influences. Although similar to sentence-completion tests, the cloze method demands deletion of random words from a passage. After administration to a group the correctly identified omissions are tallied. Experimental results show: (1) the cloze method consistently ranked three selected passages in the same way as the Flesch and Dale-Chall formulas; (2) the method was reliable; (3) the cloze method seemed to handle specialized passages more adequately than other methods; (4) the same rankings of readability were obtained when words were deleted at random or every nth word; (5) the cloze procedure could be used for comparing reading abilities of different individuals.